2 * chimeraslayercommand.cpp
5 * Created by westcott on 3/31/10.
6 * Copyright 2010 Schloss Lab. All rights reserved.
10 #include "chimeraslayercommand.h"
11 #include "chimeraslayer.h"
12 #include "deconvolutecommand.h"
14 //**********************************************************************************************************************
15 vector<string> ChimeraSlayerCommand::getValidParameters(){
17 string AlignArray[] = {"fasta", "processors","trim","split", "name","window", "include","template","numwanted", "ksize", "match","mismatch",
18 "divergence", "minsim","mincov","minbs", "minsnp","parents", "iters","outputdir","inputdir", "search","realign" };
19 vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
23 m->errorOut(e, "ChimeraSlayerCommand", "getValidParameters");
27 //**********************************************************************************************************************
28 ChimeraSlayerCommand::ChimeraSlayerCommand(){
30 abort = true; calledHelp = true;
31 vector<string> tempOutNames;
32 outputTypes["chimera"] = tempOutNames;
33 outputTypes["accnos"] = tempOutNames;
34 outputTypes["fasta"] = tempOutNames;
37 m->errorOut(e, "ChimeraSlayerCommand", "ChimeraSlayerCommand");
41 //**********************************************************************************************************************
42 vector<string> ChimeraSlayerCommand::getRequiredParameters(){
44 string AlignArray[] = {"template","fasta"};
45 vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
49 m->errorOut(e, "ChimeraSlayerCommand", "getRequiredParameters");
53 //**********************************************************************************************************************
54 vector<string> ChimeraSlayerCommand::getRequiredFiles(){
56 vector<string> myArray;
60 m->errorOut(e, "ChimeraSlayerCommand", "getRequiredFiles");
64 //***************************************************************************************************************
65 ChimeraSlayerCommand::ChimeraSlayerCommand(string option) {
67 abort = false; calledHelp = false;
69 //allow user to run help
70 if(option == "help") { help(); abort = true; calledHelp = true; }
73 //valid paramters for this command
74 string Array[] = {"fasta", "processors","name", "include","trim", "split","window", "template","numwanted", "ksize", "match","mismatch",
75 "divergence", "minsim","mincov","minbs", "minsnp","parents", "iters","outputdir","inputdir", "search","realign" };
76 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
78 OptionParser parser(option);
79 map<string,string> parameters = parser.getParameters();
81 ValidParameters validParameter("chimera.slayer");
82 map<string,string>::iterator it;
84 //check to make sure all parameters are valid for command
85 for (it = parameters.begin(); it != parameters.end(); it++) {
86 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
89 vector<string> tempOutNames;
90 outputTypes["chimera"] = tempOutNames;
91 outputTypes["accnos"] = tempOutNames;
92 outputTypes["fasta"] = tempOutNames;
94 //if the user changes the input directory command factory will send this info to us in the output parameter
95 string inputDir = validParameter.validFile(parameters, "inputdir", false);
96 if (inputDir == "not found"){ inputDir = ""; }
98 //check for required parameters
99 fastafile = validParameter.validFile(parameters, "fasta", false);
100 if (fastafile == "not found") { fastafile = ""; m->mothurOut("[ERROR]: fasta is a required parameter for the chimera.slayer command."); m->mothurOutEndLine(); abort = true; }
102 m->splitAtDash(fastafile, fastaFileNames);
104 //go through files and make sure they are good, if not, then disregard them
105 for (int i = 0; i < fastaFileNames.size(); i++) {
106 if (inputDir != "") {
107 string path = m->hasPath(fastaFileNames[i]);
108 //if the user has not given a path then, add inputdir. else leave path alone.
109 if (path == "") { fastaFileNames[i] = inputDir + fastaFileNames[i]; }
115 ableToOpen = m->openInputFile(fastaFileNames[i], in, "noerror");
117 //if you can't open it, try default location
118 if (ableToOpen == 1) {
119 if (m->getDefaultPath() != "") { //default path is set
120 string tryPath = m->getDefaultPath() + m->getSimpleName(fastaFileNames[i]);
121 m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
123 ableToOpen = m->openInputFile(tryPath, in2, "noerror");
125 fastaFileNames[i] = tryPath;
129 if (ableToOpen == 1) {
130 if (m->getOutputDir() != "") { //default path is set
131 string tryPath = m->getOutputDir() + m->getSimpleName(fastaFileNames[i]);
132 m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
134 ableToOpen = m->openInputFile(tryPath, in2, "noerror");
136 fastaFileNames[i] = tryPath;
142 if (ableToOpen == 1) {
143 m->mothurOut("Unable to open " + fastaFileNames[i] + ". It will be disregarded."); m->mothurOutEndLine();
144 //erase from file list
145 fastaFileNames.erase(fastaFileNames.begin()+i);
150 //make sure there is at least one valid file left
151 if (fastaFileNames.size() == 0) { m->mothurOut("[ERROR]: no valid files."); m->mothurOutEndLine(); abort = true; }
155 //check for required parameters
157 namefile = validParameter.validFile(parameters, "name", false);
158 if (namefile == "not found") { namefile = ""; hasName = false; }
160 m->splitAtDash(namefile, nameFileNames);
162 //go through files and make sure they are good, if not, then disregard them
163 for (int i = 0; i < nameFileNames.size(); i++) {
164 if (inputDir != "") {
165 string path = m->hasPath(nameFileNames[i]);
166 //if the user has not given a path then, add inputdir. else leave path alone.
167 if (path == "") { nameFileNames[i] = inputDir + nameFileNames[i]; }
173 ableToOpen = m->openInputFile(nameFileNames[i], in, "noerror");
175 //if you can't open it, try default location
176 if (ableToOpen == 1) {
177 if (m->getDefaultPath() != "") { //default path is set
178 string tryPath = m->getDefaultPath() + m->getSimpleName(nameFileNames[i]);
179 m->mothurOut("Unable to open " + nameFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
181 ableToOpen = m->openInputFile(tryPath, in2, "noerror");
183 nameFileNames[i] = tryPath;
187 if (ableToOpen == 1) {
188 if (m->getOutputDir() != "") { //default path is set
189 string tryPath = m->getOutputDir() + m->getSimpleName(nameFileNames[i]);
190 m->mothurOut("Unable to open " + nameFileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
192 ableToOpen = m->openInputFile(tryPath, in2, "noerror");
194 nameFileNames[i] = tryPath;
200 if (ableToOpen == 1) {
201 m->mothurOut("Unable to open " + nameFileNames[i] + ". It will be disregarded."); m->mothurOutEndLine();
202 //erase from file list
203 nameFileNames.erase(nameFileNames.begin()+i);
208 //make sure there is at least one valid file left
209 if (nameFileNames.size() == 0) { m->mothurOut("[ERROR]: no valid name files."); m->mothurOutEndLine(); abort = true; }
212 if (hasName && (nameFileNames.size() != fastaFileNames.size())) { m->mothurOut("[ERROR]: The number of namefiles does not match the number of fastafiles, please correct."); m->mothurOutEndLine(); abort=true; }
214 //if the user changes the output directory command factory will send this info to us in the output parameter
215 outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; }
219 it = parameters.find("template");
220 //user has given a template file
221 if(it != parameters.end()){
222 if (it->second == "self") { templatefile = "self"; }
224 path = m->hasPath(it->second);
225 //if the user has not given a path then, add inputdir. else leave path alone.
226 if (path == "") { parameters["template"] = inputDir + it->second; }
228 templatefile = validParameter.validFile(parameters, "template", true);
229 if (templatefile == "not open") { abort = true; }
230 else if (templatefile == "not found") { templatefile = ""; m->mothurOut("template is a required parameter for the chimera.slayer command."); m->mothurOutEndLine(); abort = true; }
234 string temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found") { temp = "1"; }
235 convert(temp, processors);
237 includeAbunds = validParameter.validFile(parameters, "include", false); if (includeAbunds == "not found") { includeAbunds = "greater"; }
238 if ((includeAbunds != "greater") && (includeAbunds != "greaterequal") && (includeAbunds != "all")) { includeAbunds = "greater"; m->mothurOut("Invalid include setting. options are greater, greaterequal or all. using greater."); m->mothurOutEndLine(); }
240 temp = validParameter.validFile(parameters, "ksize", false); if (temp == "not found") { temp = "7"; }
241 convert(temp, ksize);
243 temp = validParameter.validFile(parameters, "window", false); if (temp == "not found") { temp = "50"; }
244 convert(temp, window);
246 temp = validParameter.validFile(parameters, "match", false); if (temp == "not found") { temp = "5"; }
247 convert(temp, match);
249 temp = validParameter.validFile(parameters, "mismatch", false); if (temp == "not found") { temp = "-4"; }
250 convert(temp, mismatch);
252 temp = validParameter.validFile(parameters, "divergence", false); if (temp == "not found") { temp = "1.007"; }
255 temp = validParameter.validFile(parameters, "minsim", false); if (temp == "not found") { temp = "90"; }
256 convert(temp, minSimilarity);
258 temp = validParameter.validFile(parameters, "mincov", false); if (temp == "not found") { temp = "70"; }
259 convert(temp, minCoverage);
261 temp = validParameter.validFile(parameters, "minbs", false); if (temp == "not found") { temp = "90"; }
262 convert(temp, minBS);
264 temp = validParameter.validFile(parameters, "minsnp", false); if (temp == "not found") { temp = "100"; }
265 convert(temp, minSNP);
267 temp = validParameter.validFile(parameters, "parents", false); if (temp == "not found") { temp = "3"; }
268 convert(temp, parents);
270 temp = validParameter.validFile(parameters, "realign", false); if (temp == "not found") { temp = "f"; }
271 realign = m->isTrue(temp);
273 temp = validParameter.validFile(parameters, "trim", false); if (temp == "not found") { temp = "f"; }
274 trim = m->isTrue(temp);
276 temp = validParameter.validFile(parameters, "split", false); if (temp == "not found") { temp = "f"; }
277 trimera = m->isTrue(temp);
279 search = validParameter.validFile(parameters, "search", false); if (search == "not found") { search = "distance"; }
281 temp = validParameter.validFile(parameters, "iters", false); if (temp == "not found") { temp = "100"; }
282 convert(temp, iters);
284 temp = validParameter.validFile(parameters, "increment", false); if (temp == "not found") { temp = "5"; }
285 convert(temp, increment);
287 temp = validParameter.validFile(parameters, "numwanted", false); if (temp == "not found") { temp = "15"; }
288 convert(temp, numwanted);
290 if ((search != "distance") && (search != "blast") && (search != "kmer")) { m->mothurOut(search + " is not a valid search."); m->mothurOutEndLine(); abort = true; }
293 catch(exception& e) {
294 m->errorOut(e, "ChimeraSlayerCommand", "ChimeraSlayerCommand");
298 //**********************************************************************************************************************
300 void ChimeraSlayerCommand::help(){
303 m->mothurOut("The chimera.slayer command reads a fastafile and templatefile and outputs potentially chimeric sequences.\n");
304 m->mothurOut("This command was modeled after the chimeraSlayer written by the Broad Institute.\n");
305 m->mothurOut("The chimera.slayer command parameters are fasta, name, template, processors, trim, ksize, window, match, mismatch, divergence. minsim, mincov, minbs, minsnp, parents, search, iters, increment and numwanted.\n"); //realign,
306 m->mothurOut("The fasta parameter allows you to enter the fasta file containing your potentially chimeric sequences, and is required. \n");
307 m->mothurOut("The name parameter allows you to provide a name file, if you are using template=self. \n");
308 m->mothurOut("You may enter multiple fasta files by separating their names with dashes. ie. fasta=abrecovery.fasta-amazon.fasta \n");
309 m->mothurOut("The template parameter allows you to enter a template file containing known non-chimeric sequences, and is required. You may also set template=self, in this case the abundant sequences will be used as potential parents. \n");
310 m->mothurOut("The processors parameter allows you to specify how many processors you would like to use. The default is 1. \n");
312 m->mothurOut("When using MPI, the processors parameter is set to the number of MPI processes running. \n");
314 m->mothurOut("The trim parameter allows you to output a new fasta file containing your sequences with the chimeric ones trimmed to include only their longest piece, default=F. \n");
315 m->mothurOut("The split parameter allows you to check both pieces of non-chimeric sequence for chimeras, thus looking for trimeras and quadmeras. default=F. \n");
316 m->mothurOut("The window parameter allows you to specify the window size for searching for chimeras, default=50. \n");
317 m->mothurOut("The increment parameter allows you to specify how far you move each window while finding chimeric sequences, default=5.\n");
318 m->mothurOut("The numwanted parameter allows you to specify how many sequences you would each query sequence compared with, default=15.\n");
319 m->mothurOut("The ksize parameter allows you to input kmersize, default is 7, used if search is kmer. \n");
320 m->mothurOut("The match parameter allows you to reward matched bases in blast search, default is 5. \n");
321 m->mothurOut("The parents parameter allows you to select the number of potential parents to investigate from the numwanted best matches after rating them, default is 3. \n");
322 m->mothurOut("The mismatch parameter allows you to penalize mismatched bases in blast search, default is -4. \n");
323 m->mothurOut("The divergence parameter allows you to set a cutoff for chimera determination, default is 1.007. \n");
324 m->mothurOut("The iters parameter allows you to specify the number of bootstrap iters to do with the chimeraslayer method, default=100.\n");
325 m->mothurOut("The minsim parameter allows you to specify a minimum similarity with the parent fragments, default=90. \n");
326 m->mothurOut("The mincov parameter allows you to specify minimum coverage by closest matches found in template. Default is 70, meaning 70%. \n");
327 m->mothurOut("The minbs parameter allows you to specify minimum bootstrap support for calling a sequence chimeric. Default is 90, meaning 90%. \n");
328 m->mothurOut("The minsnp parameter allows you to specify percent of SNPs to sample on each side of breakpoint for computing bootstrap support (default: 100) \n");
329 m->mothurOut("The search parameter allows you to specify search method for finding the closest parent. Choices are distance, blast, and kmer, default distance. \n");
330 m->mothurOut("The realign parameter allows you to realign the query to the potential parents. Choices are true or false, default false. \n");
331 m->mothurOut("The chimera.slayer command should be in the following format: \n");
332 m->mothurOut("chimera.slayer(fasta=yourFastaFile, template=yourTemplate, search=yourSearch) \n");
333 m->mothurOut("Example: chimera.slayer(fasta=AD.align, template=core_set_aligned.imputed.fasta, search=kmer) \n");
334 m->mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFastaFile).\n\n");
336 catch(exception& e) {
337 m->errorOut(e, "ChimeraSlayerCommand", "help");
342 //***************************************************************************************************************
344 ChimeraSlayerCommand::~ChimeraSlayerCommand(){ /* do nothing */ }
346 //***************************************************************************************************************
348 int ChimeraSlayerCommand::execute(){
351 if (abort == true) { if (calledHelp) { return 0; } return 2; }
353 for (int s = 0; s < fastaFileNames.size(); s++) {
355 m->mothurOut("Checking sequences from " + fastaFileNames[s] + " ..." ); m->mothurOutEndLine();
357 int start = time(NULL);
359 if (templatefile != "self") { //you want to run slayer with a refernce template
360 chimera = new ChimeraSlayer(fastaFileNames[s], templatefile, trim, search, ksize, match, mismatch, window, divR, minSimilarity, minCoverage, minBS, minSNP, parents, iters, increment, numwanted, realign);
362 if (nameFileNames.size() != 0) { //you provided a namefile and we don't need to create one
363 chimera = new ChimeraSlayer(fastaFileNames[s], templatefile, trim, nameFileNames[s], search, includeAbunds, ksize, match, mismatch, window, divR, minSimilarity, minCoverage, minBS, minSNP, parents, iters, increment, numwanted, realign);
366 m->mothurOutEndLine(); m->mothurOut("No namesfile given, running unique.seqs command to generate one."); m->mothurOutEndLine(); m->mothurOutEndLine();
368 //use unique.seqs to create new name and fastafile
369 string inputString = "fasta=" + fastaFileNames[s];
370 m->mothurOut("/******************************************/"); m->mothurOutEndLine();
371 m->mothurOut("Running command: unique.seqs(" + inputString + ")"); m->mothurOutEndLine();
373 Command* uniqueCommand = new DeconvoluteCommand(inputString);
374 uniqueCommand->execute();
376 map<string, vector<string> > filenames = uniqueCommand->getOutputFiles();
378 delete uniqueCommand;
380 m->mothurOut("/******************************************/"); m->mothurOutEndLine();
382 string nameFile = filenames["name"][0];
383 fastaFileNames[s] = filenames["fasta"][0];
385 chimera = new ChimeraSlayer(fastaFileNames[s], templatefile, trim, nameFile, search, includeAbunds, ksize, match, mismatch, window, divR, minSimilarity, minCoverage, minBS, minSNP, parents, iters, increment, numwanted, realign);
389 if (outputDir == "") { outputDir = m->hasPath(fastaFileNames[s]); }//if user entered a file with a path then preserve it
390 string outputFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "slayer.chimera";
391 string accnosFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "slayer.accnos";
392 string trimFastaFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "slayer.fasta";
394 if (m->control_pressed) { delete chimera; for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } return 0; }
396 if (chimera->getUnaligned()) {
397 m->mothurOut("Your template sequences are different lengths, please correct."); m->mothurOutEndLine();
401 templateSeqsLength = chimera->getLength();
404 int pid, numSeqsPerProcessor;
406 vector<unsigned long int> MPIPos;
409 MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
410 MPI_Comm_size(MPI_COMM_WORLD, &processors);
414 MPI_File outMPIAccnos;
415 MPI_File outMPIFasta;
417 int outMode=MPI_MODE_CREATE|MPI_MODE_WRONLY;
418 int inMode=MPI_MODE_RDONLY;
420 char outFilename[1024];
421 strcpy(outFilename, outputFileName.c_str());
423 char outAccnosFilename[1024];
424 strcpy(outAccnosFilename, accnosFileName.c_str());
426 char outFastaFilename[1024];
427 strcpy(outFastaFilename, trimFastaFileName.c_str());
429 char inFileName[1024];
430 strcpy(inFileName, fastaFileNames[s].c_str());
432 MPI_File_open(MPI_COMM_WORLD, inFileName, inMode, MPI_INFO_NULL, &inMPI); //comm, filename, mode, info, filepointer
433 MPI_File_open(MPI_COMM_WORLD, outFilename, outMode, MPI_INFO_NULL, &outMPI);
434 MPI_File_open(MPI_COMM_WORLD, outAccnosFilename, outMode, MPI_INFO_NULL, &outMPIAccnos);
435 if (trim) { MPI_File_open(MPI_COMM_WORLD, outFastaFilename, outMode, MPI_INFO_NULL, &outMPIFasta); }
437 if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&inMPI); MPI_File_close(&outMPI); if (trim) { MPI_File_close(&outMPIFasta); } MPI_File_close(&outMPIAccnos); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } delete chimera; return 0; }
439 if (pid == 0) { //you are the root process
440 m->mothurOutEndLine();
441 m->mothurOut("Only reporting sequence supported by " + toString(minBS) + "% of bootstrapped results.");
442 m->mothurOutEndLine();
444 string outTemp = "Name\tLeftParent\tRightParent\tDivQLAQRB\tPerIDQLAQRB\tBootStrapA\tDivQLBQRA\tPerIDQLBQRA\tBootStrapB\tFlag\tLeftWindow\tRightWindow\n";
447 int length = outTemp.length();
448 char* buf2 = new char[length];
449 memcpy(buf2, outTemp.c_str(), length);
451 MPI_File_write_shared(outMPI, buf2, length, MPI_CHAR, &status);
454 MPIPos = m->setFilePosFasta(fastaFileNames[s], numSeqs); //fills MPIPos, returns numSeqs
456 //send file positions to all processes
457 for(int i = 1; i < processors; i++) {
458 MPI_Send(&numSeqs, 1, MPI_INT, i, tag, MPI_COMM_WORLD);
459 MPI_Send(&MPIPos[0], (numSeqs+1), MPI_LONG, i, tag, MPI_COMM_WORLD);
462 //figure out how many sequences you have to align
463 numSeqsPerProcessor = numSeqs / processors;
464 int startIndex = pid * numSeqsPerProcessor;
465 if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; }
468 driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, outMPIFasta, MPIPos);
470 if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&inMPI); MPI_File_close(&outMPI); if (trim) { MPI_File_close(&outMPIFasta); } MPI_File_close(&outMPIAccnos); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } remove(outputFileName.c_str()); remove(accnosFileName.c_str()); delete chimera; return 0; }
472 }else{ //you are a child process
473 MPI_Recv(&numSeqs, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status);
474 MPIPos.resize(numSeqs+1);
475 MPI_Recv(&MPIPos[0], (numSeqs+1), MPI_LONG, 0, tag, MPI_COMM_WORLD, &status);
477 //figure out how many sequences you have to align
478 numSeqsPerProcessor = numSeqs / processors;
479 int startIndex = pid * numSeqsPerProcessor;
480 if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; }
483 driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, outMPIFasta, MPIPos);
485 if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&inMPI); MPI_File_close(&outMPI); if (trim) { MPI_File_close(&outMPIFasta); } MPI_File_close(&outMPIAccnos); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } delete chimera; return 0; }
489 MPI_File_close(&inMPI);
490 MPI_File_close(&outMPI);
491 MPI_File_close(&outMPIAccnos);
492 if (trim) { MPI_File_close(&outMPIFasta); }
493 MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case
497 string tempHeader = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "slayer.chimeras.tempHeader";
498 m->openOutputFile(tempHeader, outHeader);
500 chimera->printHeader(outHeader);
503 vector<unsigned long int> positions = m->divideFile(fastaFileNames[s], processors);
505 for (int i = 0; i < (positions.size()-1); i++) {
506 lines.push_back(new linePair(positions[i], positions[(i+1)]));
510 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
512 numSeqs = driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName, trimFastaFileName);
514 if (m->control_pressed) { outputTypes.clear(); if (trim) { remove(trimFastaFileName.c_str()); } remove(outputFileName.c_str()); remove(tempHeader.c_str()); remove(accnosFileName.c_str()); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); delete chimera; return 0; }
517 processIDS.resize(0);
519 numSeqs = createProcesses(outputFileName, fastaFileNames[s], accnosFileName, trimFastaFileName);
521 rename((outputFileName + toString(processIDS[0]) + ".temp").c_str(), outputFileName.c_str());
522 rename((accnosFileName + toString(processIDS[0]) + ".temp").c_str(), accnosFileName.c_str());
523 if (trim) { rename((trimFastaFileName + toString(processIDS[0]) + ".temp").c_str(), trimFastaFileName.c_str()); }
525 //append output files
526 for(int i=1;i<processors;i++){
527 m->appendFiles((outputFileName + toString(processIDS[i]) + ".temp"), outputFileName);
528 remove((outputFileName + toString(processIDS[i]) + ".temp").c_str());
531 //append output files
532 for(int i=1;i<processors;i++){
533 m->appendFiles((accnosFileName + toString(processIDS[i]) + ".temp"), accnosFileName);
534 remove((accnosFileName + toString(processIDS[i]) + ".temp").c_str());
538 for(int i=1;i<processors;i++){
539 m->appendFiles((trimFastaFileName + toString(processIDS[i]) + ".temp"), trimFastaFileName);
540 remove((trimFastaFileName + toString(processIDS[i]) + ".temp").c_str());
544 if (m->control_pressed) { outputTypes.clear(); if (trim) { remove(trimFastaFileName.c_str()); } remove(outputFileName.c_str()); remove(accnosFileName.c_str()); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); delete chimera; return 0; }
548 numSeqs = driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName, trimFastaFileName);
550 if (m->control_pressed) { outputTypes.clear(); if (trim) { remove(trimFastaFileName.c_str()); } remove(outputFileName.c_str()); remove(tempHeader.c_str()); remove(accnosFileName.c_str()); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); delete chimera; return 0; }
554 m->appendFiles(outputFileName, tempHeader);
556 remove(outputFileName.c_str());
557 rename(tempHeader.c_str(), outputFileName.c_str());
563 for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear();
565 outputNames.push_back(outputFileName); outputTypes["chimera"].push_back(outputFileName);
566 outputNames.push_back(accnosFileName); outputTypes["accnos"].push_back(accnosFileName);
567 if (trim) { outputNames.push_back(trimFastaFileName); outputTypes["fasta"].push_back(trimFastaFileName); }
569 m->mothurOutEndLine(); m->mothurOut("It took " + toString(time(NULL) - start) + " secs to check " + toString(numSeqs) + " sequences."); m->mothurOutEndLine();
572 //set accnos file as new current accnosfile
574 itTypes = outputTypes.find("accnos");
575 if (itTypes != outputTypes.end()) {
576 if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setAccnosFile(current); }
580 itTypes = outputTypes.find("fasta");
581 if (itTypes != outputTypes.end()) {
582 if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setFastaFile(current); }
586 m->mothurOutEndLine();
587 m->mothurOut("Output File Names: "); m->mothurOutEndLine();
588 for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
589 m->mothurOutEndLine();
594 catch(exception& e) {
595 m->errorOut(e, "ChimeraSlayerCommand", "execute");
599 //**********************************************************************************************************************
601 int ChimeraSlayerCommand::driver(linePair* filePos, string outputFName, string filename, string accnos, string fasta){
604 m->openOutputFile(outputFName, out);
607 m->openOutputFile(accnos, out2);
610 if (trim) { m->openOutputFile(fasta, out3); }
613 m->openInputFile(filename, inFASTA);
615 inFASTA.seekg(filePos->start);
622 if (m->control_pressed) { out.close(); out2.close(); if (trim) { out3.close(); } inFASTA.close(); return 1; }
624 Sequence* candidateSeq = new Sequence(inFASTA); m->gobble(inFASTA);
625 string candidateAligned = candidateSeq->getAligned();
627 if (candidateSeq->getName() != "") { //incase there is a commented sequence at the end of a file
629 if (candidateSeq->getAligned().length() != templateSeqsLength) {
630 m->mothurOut(candidateSeq->getName() + " is not the same length as the template sequences. Skipping."); m->mothurOutEndLine();
633 chimera->getChimeras(candidateSeq);
635 if (m->control_pressed) { delete candidateSeq; return 1; }
637 //if you are not chimeric, then check each half
638 data_results wholeResults = chimera->getResults();
640 //determine if we need to split
641 bool isChimeric = false;
643 if (wholeResults.flag == "yes") {
644 string chimeraFlag = "no";
645 if( (wholeResults.results[0].bsa >= minBS && wholeResults.results[0].divr_qla_qrb >= divR)
647 (wholeResults.results[0].bsb >= minBS && wholeResults.results[0].divr_qlb_qra >= divR) ) { chimeraFlag = "yes"; }
650 if (chimeraFlag == "yes") {
651 if ((wholeResults.results[0].bsa >= minBS) || (wholeResults.results[0].bsb >= minBS)) { isChimeric = true; }
655 if ((!isChimeric) && trimera) {
657 //split sequence in half by bases
658 string leftQuery, rightQuery;
659 Sequence tempSeq(candidateSeq->getName(), candidateAligned);
660 divideInHalf(tempSeq, leftQuery, rightQuery);
662 //run chimeraSlayer on each piece
663 Sequence* left = new Sequence(candidateSeq->getName(), leftQuery);
664 Sequence* right = new Sequence(candidateSeq->getName(), rightQuery);
667 chimera->getChimeras(left);
668 data_results leftResults = chimera->getResults();
670 chimera->getChimeras(right);
671 data_results rightResults = chimera->getResults();
673 //if either piece is chimeric then report
674 Sequence* trimmed = chimera->print(out, out2, leftResults, rightResults);
675 if (trim) { trimmed->printSequence(out3); delete trimmed; }
677 delete left; delete right;
679 }else { //already chimeric
681 Sequence* trimmed = chimera->print(out, out2);
682 if (trim) { trimmed->printSequence(out3); delete trimmed; }
691 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
692 unsigned long int pos = inFASTA.tellg();
693 if ((pos == -1) || (pos >= filePos->end)) { break; }
695 if (inFASTA.eof()) { break; }
699 if((count) % 100 == 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine(); }
702 if((count) % 100 != 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine(); }
706 if (trim) { out3.close(); }
711 catch(exception& e) {
712 m->errorOut(e, "ChimeraSlayerCommand", "driver");
716 //**********************************************************************************************************************
718 int ChimeraSlayerCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& outMPI, MPI_File& outAccMPI, MPI_File& outFastaMPI, vector<unsigned long int>& MPIPos){
722 MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
724 for(int i=0;i<num;i++){
726 if (m->control_pressed) { return 1; }
729 int length = MPIPos[start+i+1] - MPIPos[start+i];
731 char* buf4 = new char[length];
732 MPI_File_read_at(inMPI, MPIPos[start+i], buf4, length, MPI_CHAR, &status);
734 string tempBuf = buf4;
735 if (tempBuf.length() > length) { tempBuf = tempBuf.substr(0, length); }
736 istringstream iss (tempBuf,istringstream::in);
740 Sequence* candidateSeq = new Sequence(iss); m->gobble(iss);
741 string candidateAligned = candidateSeq->getAligned();
743 if (candidateSeq->getName() != "") { //incase there is a commented sequence at the end of a file
745 if (candidateSeq->getAligned().length() != templateSeqsLength) {
746 m->mothurOut(candidateSeq->getName() + " is not the same length as the template sequences. Skipping."); m->mothurOutEndLine();
750 chimera->getChimeras(candidateSeq);
752 if (m->control_pressed) { delete candidateSeq; return 1; }
754 //if you are not chimeric, then check each half
755 data_results wholeResults = chimera->getResults();
757 //determine if we need to split
758 bool isChimeric = false;
760 if (wholeResults.flag == "yes") {
761 string chimeraFlag = "no";
762 if( (wholeResults.results[0].bsa >= minBS && wholeResults.results[0].divr_qla_qrb >= divR)
764 (wholeResults.results[0].bsb >= minBS && wholeResults.results[0].divr_qlb_qra >= divR) ) { chimeraFlag = "yes"; }
767 if (chimeraFlag == "yes") {
768 if ((wholeResults.results[0].bsa >= minBS) || (wholeResults.results[0].bsb >= minBS)) { isChimeric = true; }
772 if ((!isChimeric) && trimera) {
773 //split sequence in half by bases
774 string leftQuery, rightQuery;
775 Sequence tempSeq(candidateSeq->getName(), candidateAligned);
776 divideInHalf(tempSeq, leftQuery, rightQuery);
778 //run chimeraSlayer on each piece
779 Sequence* left = new Sequence(candidateSeq->getName(), leftQuery);
780 Sequence* right = new Sequence(candidateSeq->getName(), rightQuery);
783 chimera->getChimeras(left);
784 data_results leftResults = chimera->getResults();
786 chimera->getChimeras(right);
787 data_results rightResults = chimera->getResults();
789 //if either piece is chimeric then report
790 Sequence* trimmed = chimera->print(outMPI, outAccMPI, leftResults, rightResults);
792 string outputString = ">" + trimmed->getName() + "\n" + trimmed->getAligned() + "\n";
795 //write to accnos file
796 int length = outputString.length();
797 char* buf2 = new char[length];
798 memcpy(buf2, outputString.c_str(), length);
800 MPI_File_write_shared(outFastaMPI, buf2, length, MPI_CHAR, &status);
804 delete left; delete right;
808 Sequence* trimmed = chimera->print(outMPI, outAccMPI);
811 string outputString = ">" + trimmed->getName() + "\n" + trimmed->getAligned() + "\n";
814 //write to accnos file
815 int length = outputString.length();
816 char* buf2 = new char[length];
817 memcpy(buf2, outputString.c_str(), length);
819 MPI_File_write_shared(outFastaMPI, buf2, length, MPI_CHAR, &status);
829 if((i+1) % 100 == 0){ cout << "Processing sequence: " << (i+1) << endl; m->mothurOutJustToLog("Processing sequence: " + toString(i+1) + "\n"); }
832 if(num % 100 != 0){ cout << "Processing sequence: " << num << endl; m->mothurOutJustToLog("Processing sequence: " + toString(num) + "\n"); }
837 catch(exception& e) {
838 m->errorOut(e, "ChimeraSlayerCommand", "driverMPI");
844 /**************************************************************************************************/
846 int ChimeraSlayerCommand::createProcesses(string outputFileName, string filename, string accnos, string fasta) {
848 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
852 //loop through and create all the processes you want
853 while (process != processors) {
857 processIDS.push_back(pid); //create map from line number to pid so you can append files in correct order later
860 num = driver(lines[process], outputFileName + toString(getpid()) + ".temp", filename, accnos + toString(getpid()) + ".temp", fasta + toString(getpid()) + ".temp");
862 //pass numSeqs to parent
864 string tempFile = outputFileName + toString(getpid()) + ".num.temp";
865 m->openOutputFile(tempFile, out);
871 m->mothurOut("[ERROR]: unable to spawn the necessary processes."); m->mothurOutEndLine();
872 for (int i = 0; i < processIDS.size(); i++) { kill (processIDS[i], SIGINT); }
877 //force parent to wait until all the processes are done
878 for (int i=0;i<processors;i++) {
879 int temp = processIDS[i];
883 for (int i = 0; i < processIDS.size(); i++) {
885 string tempFile = outputFileName + toString(processIDS[i]) + ".num.temp";
886 m->openInputFile(tempFile, in);
887 if (!in.eof()) { int tempNum = 0; in >> tempNum; num += tempNum; }
888 in.close(); remove(tempFile.c_str());
894 catch(exception& e) {
895 m->errorOut(e, "ChimeraSlayerCommand", "createProcesses");
900 /**************************************************************************************************/
902 int ChimeraSlayerCommand::divideInHalf(Sequence querySeq, string& leftQuery, string& rightQuery) {
905 string queryUnAligned = querySeq.getUnaligned();
906 int numBases = int(queryUnAligned.length() * 0.5);
908 string queryAligned = querySeq.getAligned();
909 leftQuery = querySeq.getAligned();
910 rightQuery = querySeq.getAligned();
914 for (int i = 0; i < queryAligned.length(); i++) {
916 if (isalpha(queryAligned[i])) {
921 if (baseCount >= numBases) { leftSpot = i; break; } //first half
924 //blank out right side
925 for (int i = leftSpot; i < leftQuery.length(); i++) { leftQuery[i] = '.'; }
927 //blank out left side
928 for (int i = 0; i < leftSpot; i++) { rightQuery[i] = '.'; }
933 catch(exception& e) {
934 m->errorOut(e, "ChimeraSlayerCommand", "divideInHalf");
939 /**************************************************************************************************/